Intérieur Techniques de localisation pour les réseaux de capteurs sans fil

In this thesis, the author focused on RSSI basedlocalization algorithms for indoor applications inwireless sensor networks. Firstly, from the observationof RSSI behavior based on an experimentallocalization system, an experimental RSSI channelmodel is deduced, which is consistent to the popularlognormal shadowing path loss model. Secondly, thisthesis proposes three indoor localization algorithmsbased on multilateration and averaged RSSI. In thesealgorithms, the measured distances are weightedaccording to their assumed accuracy. Lastly, a RSSIbased parameter tracking strategy for constrainedposition localization is proposed. To estimate channelmodel parameters, least mean squares method (LMS)is associated with the trilateration method.Quantitative criteria are provided to guarantee theefficiency of the proposed tracking strategy byproviding a tradeoff between the constraint resolutionand parameter variation. The simulation results showa good behavior of the proposed tracking strategy inpresence of space-time variation of the propagationchannel. Compared with the existing RSSI basedalgorithms, the proposed tracking strategy exhibitsbetter localization accuracy but consumes morecalculation time. In addition, experimental tracking testis performed to validate the effectiveness of theproposed tracking strategy